Social Media Analytics

• Sponsored by: Capgemini
• Scientific Lead: M.Sc Matthias Wissel
• Project Lead: Dr. Ricardo Acevedo Cabra
• Term: Summer semester 2018

Social media platforms more and more define how we consume news and also provide the place for political and social discourse. The forming of opinions on and the change of mass perception of certain topics occurs a lot faster than in the classical media setting. Therefore the analysis of social media before and after special events became crucial to understanding what drove the decisions that were made for example in a public poll. To get to know the methodologies to do this the aim for the project is to use techniques from natural language processing (NLP), time series analysis and classification to gain insights from social media data. These insights will come from the meta data of the posts as well as the concrete content of the messages. Also, the analysis of the network structure will be an important issue. We will go through all the stations necessary to gather these insights, from general exploration of the data, over finding additional secondary data sources to deploying the models and visualizing the results. Since the size of the data to be analyzed is too big for single machines we will use distributed frameworks like spark for the data processing and machine learning algorithms.

Results: The results of this project were summarised in the final presentation and explained in detail in the final documentation.